With the advent of robotic placement devices, the number of products produced in automated factories have increased many folds within the past decade. This has culminated in most manufacturers and system designers feverishly focusing on further improvements to robotic placement devices to get an advantage in manufacturing efficiency. Typically, manufacturers are faced with the problem of trying to place uniquely-shaped components into areas or apertures with extremely close tolerances. Regrettable, conventional robotic automated placement devices, after securing the component to be placed, must identify its orientation before it can determine a three-dimensional offset to effectuate proper component placement.
One such method currently used involves taking a series of electronic images (pictures) to determine a fixed location from which an offset position may be calculated for the component placement. However, with this method, the more uniquely-shaped components require an extreme number of images to be taken to properly locate the component in a three-dimensional space. Unfortunately, this method requires the use of expensive equipment at a plurality of workstations, which is a very costly alternative. Furthermore, taking several images may involve many rotations (or other movement) of the component to be placed resulting in an unreasonable long delay in determining the component orientation.
Another known method involves rotating the components to produce shadows of the component that are used to determine a three-dimensional orientation. This method eliminates the use of expensive equipment, such as, cameras for providing images. Regrettable, however, a similar number of rotations are required to produce enough different shadow views to properly determine component location, resulting in the approximately the same time delays involved in the previous method.
Thus, what is needed is a automated robotic system that determines a three-dimensional location of a component to be placed quickly and economically.